The use of computer Technologies -Sakineh Yeganegi
Use of Computer in Testing ' ' ' ' Development in computer technology had a main effect on testing, important language test such as TOEFL are moving to computer based testing (CBT).The propose of computer based testing make out standing background, application of CATs, some advantages and disadvantages and who can use computer adaptive tests and what are the limitation of computer adaptive test in testing? Test selection and the result depend on the continuing pattern of success and failure which experience by the test taker. If the test taker is succeeds on given item, another difficult item is present, and if first experience failure then easier item is presented. In this way testing happen until enough information gather to assess special test taker ability. Second language computer adaptive testing (CAT) is a technologically advanced method of assessment in which the computer selects and presents test items to examinees according to the estimated level of the examinee’s language ability. In 1960s and 1970s the U.S Department of Defence notice the benefits of adaptive testing. Early attempts to build adaptive tests by the U.S Army, Navy, and Air Force were often less than successful, very expensive, and used large- scale computers. By the early 1980s personal computers had acquired the power of large-scale and in late 1980s, CAT used in Colleage Board’s CAT Graduate Record Examination and with the work of in-house researchers in foreign language education at the universities in United State, Britain and other countries. Today, with development of software companies assisting test developers with their institutional second language CATs, computer adaptive testing finally become a viable alternative to conventional paper- and- pencil testing. (Mousavi, p.131). Computer adaptive test (CATs) can be used in different purposes including the following: - Identifying whether an individual has met the specific objectives of a basic language or literature course. - Indicating an individual’s level of achievement in a skill domain. - Identifying specific areas in which a student needs additional educational experiences - Diagnosing an individual’s skill-area strengths - Detecting whether candidates have met minimum course requirements as demonstrated in mastery test. In CAT individuals take a test that is specific to his or her ability level. CAT was developed to identify the time consuming and inefficient test present easy questions to high-ability persons and properly difficult questions to low-ability testees. Advantages of using computers in language testing includes in two categories: a) testing considerations, b)'' human considerations.'' a) testing considerations are as follows: - Computers are much more accurate at scoring selected –response test than human beings - Computers are more accurate at reporting scores. - Computers can give immediate feedback in the form of a report test scores, complete with a printout of basic testing statistics. - IRT and CAT allow testes to target the specific ability levels of individual students that can provide more precise estimates of those abilities. - The use of different test for each student should minimize any practice effects, studying for the test and CHEATING. - Diagnostic feedback can be provided very quickly to each student on those items answered incorrectly if that is the purpose of the test, this feedback can be descriptive if artificial intelligence is used. - CATs take less time to finish than traditional paper-and- pencil test and test efficiency improve and substantially reducing time for each exam. - Test can include text, graphics, photographs, and even full-motion video clips. - CATs have the benefit of accurate, consistent evaluation. - CATs provide teachers and test administrators with diagnostic assistance. - There is a swift access to item bank. - There is flexibility in scheduling test since each administration is unique exam. - Test can be linked and used in a network. a) human considerations are as follows: - Use of computers let the students to work at their speed to cope with exam. - In CATs student should experience less disappointment than on paper-and-pencil test, because they are working on the item test that are appropriate for their level of abilities. - Students may understand that CATs are less overwhelming because the questions are presented one at a time on the screen rather than in an intimidating test booklet with hundreds of test items. - Many students like computers and enjoy the testing process. - It is more convenient to provide exam results immediately. - In CATs test security is improved . - The use of CATs results in a relief to test writers. - The problem of students handwriting is eliminated. - CATs are superior in termes of reliability and validity rather than paper-and- pencil test when few items are administered. - The combined CAT and non-CAT measurement result in an excellent predictive power. - The use of CAT results in improved measurement precision. It also involves test free person measurement and sample-free item measurement. - Item are specific to the ability level of the examinee. - The use of CATs minimizes disappointment and weariness of students in response to questions of items that beyond their abilities. - It reduces the boredom for those examinees who would otherwise be asked to encounter a number of items that would be too easy for them. Computerized testing developments have more benefits for agencies and groups to use CATs except test takers includes as follow: #Educators considering the use of a published or in-house CAT to assess student achievement in large-enrolment second language classrooms or programs. #Licensing boards needing to develop a CAT to help them identify candidates who met specific performance standards for giving permit. #Agencies preparing user guides for their computer adaptive achievement test such as ETSs . #Department of Education wishing to develop a CAT version of state-wide minimum competency tests. #Departments of modern foreign languages wanting to create a proficiency CAT for entrance into or exit from required language courses. In use of CATs there are limitations of computerize test in language testing, so the disadvantages of using computers in language testing subdivided into two categories: a) Physical consideration and b) Performance considerations: a) Physical considerations ''includes as follow: '' - The high cost of computer hardware and software mat pose potential problems for CAT users. Computer equipment may not always be available, or in working order. Reliable sources of electricity are not universally available. - Screen capacity is another physical consideration. While most computers today have overcome the 80 characters by 25 line restrictions of a few years ago, the amount of material can be presented in the screen is limited. - The graphic capabilities of many computers may be limited. - There is the potential of double jeopardy. b) Performance considerations includes as follow: - The presentation of a test on computer may lead to different results from those hat prepare in the format of paper and pencil test. - Difference degree of student’s ability in using computer may lead discrepancy in their performances on computer. - There is the potential of subjecting test candidates to establishing level of anxiety as a result of the need to manipulate the new technology, the potential bias in computerized exam against those familiar with new technology. - Most CAT items are limited to only objective recognition format. - CAT approaches fail to introduce a warm-up at the beginning of the testing session. - There is a possibility of legal objections to the promulgations of scores for different students based on different combinations of items which are used for the identical academic decisions. Also, there are four basic categories of computer adaptive tests: 1. Decision point tests, 2. Step Ladder test, 3. Error- controlled Tests, 4. Multi-Stage Tests 1. Decision point Test: This test keeps the capability of shaping information function of the test. Items in the bank are limited to difficulty points where decision of admission, examination must be made what happen to this test. If specific part of ability has been decided for admission to or exemption from a program of instruction then the computer presents the candidate only with items at the corresponding difficulty level, rather than with items at all points along the difficulty continuum. The cut –off decision is made by the computer algorithm at some prespecified level of acceptable error. This kind of program tends to be highly efficient for decision- making purpose in that highly accurate decisions may be made on the basis of responses to very few items. 2. Step Ladder Tests: This kind of test is constructed in order to specify number of items is held in the bank at each of a specified number of proficiency or achievement steps. This assume that all the items have been pre-analyzed, to test, and arranged in rank order on difficulty continuum. Example, computer adaptive test with 500 items arranged so that 10 items appear at each of 50 difficulty steps. The computer algorithm would select some entry level for a given examinee. Based on the success or failure with a given item, a more appropriate item would be selected at the specified number of steps above or below the first item. 3. Error-Controlled Tests: Type of CATs that employs a procedure such as unconditional maximum like hood estimation in order to estimate ability on the ability continuum. These test access and present the item in a bank that is nearest in difficulty to the estimated person ability. After each new item is encountered, a revised estimate of person ability is provided with an associated estimate of measurement standard error. 4. Multi-Stage Tests: These type of CATs bases selection of successive sub-tests to be attempted on performance with prior subtests already attempted. A decision is made on the basis of performance on the first stage test that determines which of several tests of different difficulty will be administered during the second stage. This type of test minimizes the number of items needed for accurate decision, and maximizes Item Information Function of test. ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '